Fantastic report on the state of AI in grassroots and non-profit organisations, including how they’re using it, why, governance structures, the differences between large and small organisations, and the value that they can have in a wider debate.

This paper provides a deeper dive into using AI in your organisation — balancing techno-determinism (that AI is an inevitability), how the general economic downturn affects decisions re AI,

Research Questions:

• Why and how do non-profit and grassroots organisations engage with generative AI tools?
• What are the main drivers and key elements that shape this engagement?
• How do these organisations see their role in shaping the broader AI debate?

Key Findings

• [As many as] 78% of non-profit and grassroots organisations use generative AI tools in some capacity
• 73% [minimum] of organisations do not have policies or guidelines about AI in place.
  • Of those organisations with AI policies or guidelines, two thirds have annual incomes exceeding £1 million.
  • Smaller organisations, particularly those without existing exposure to new technologies, or high levels of digital literacy, or whose values conflict with generative AI usage (such as climate concerns), either avoid using these tools, or do so without proper governance structures.
• [At least] 59% of organisations are not engaged in the broader AI debate, citing a lack of awareness of opportunities for involvement, not being asked to get involved, and resource limitations as the main factors.
  • Some recognise the sector’s unique value in being well placed to advocate for communities at a time of significant change. However, there is limited consensus on its role, and little mention of the mechanisms at its disposal to co-ordinate efforts, hold government and industry to account, or influence decisions.
• NOTE: Some organisations declined to participate, citing no generative AI use and uncertainty about how they could then contribute to the research. The lower rate of non-generative AI users choosing to engage in the research may have disproportionately skewed the findings of high levels of use in the sector.

Methodology

• Unlike Charity Digital Skills Report 2024 and CAST’s AI survey, uses more qualitative methods
• Focuses on NFP with social/enviro missions’ approach to GenAI, and engagement with wider AI discourse. Explores economic realities of orgs, power dynamics, etc
• All data was collected between February and April 2024.
• Survey = 51 different organisations
• Online discussions = 15 organisations
• 5 interviews

Background

• “To date, wider civil society has struggled to find its voice in discourse regarding AI and has therefore had very little influence in the UK’s AI landscape and developments.“
  • Bringing in voices from these movements (esp those not tech focused) could widen discourse
  • NFPs hold gov and industry to account, so they are needed in this space to bring in their critiques